Prediction of Discharge Functional Independence Measure (FIM) Scores in Stroke Patients in a Convalescent Rehabilitation Ward: Association Between Multidisciplinary Conference-Based Predictions and an Existing Formula-Based Prediction Model

Prediction of Discharge Functional Independence Measure (FIM) Scores in Stroke Patients in a Convalescent Rehabilitation Ward: Association Between Multidisciplinary Conference-Based Predictions and an Existing Formula-Based Prediction Model - Prediction of Discharge FIM Scores in Stroke Patients in a Convalescent Rehabilitation Ward: Association Between Multidisciplinary Conference-Based Predictions and an Existing Formula-Based Prediction Model

Status

Active, not recruiting

Phases

Unknown

Study type

Observational

Source

JPRN

Registry ID

JPRN-UMIN000060660

Enrollment

600

Registered

2026-02-12

Start date

2026-01-29

Completion date

Unknown

Last updated

2026-06-29

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Stroke

Interventions

None listed

Eligibility

Sex/Gender

All

Inclusion criteria

Inclusion criteria: The participants will be stroke patients admitted to the convalescent rehabilitation ward of Morinomiya Hospital between March 2023 and May 2025.

Exclusion criteria

Exclusion criteria: Recurrent cases and patients who experienced clinical deterioration during hospitalization and were transferred to another ward or hospital, or who died, were excluded. In addition, because the existing formula-based model used as a comparator in this study was developed primarily for patients with cerebral infarction and intracerebral hemorrhage, cases of subarachnoid hemorrhage were also excluded from the analysis to ensure a fair comparison and a homogeneous study population. Cases with missing data were likewise excluded.

Design outcomes

Primary

Measure	Time frame
MAE of discharge mFIM prediction(conference vs formula)	—

Secondary

Measure	Time frame
RMSE,Prediction bias (predicted minus observed),R2,Concordance correlation coefficient (CCC),Calibration intercept and slope,Bland-Altman limits of agreement (LoA),Proportion within error (plus/minus 5/10/15 points),Subgroup performance by admission mFIM severity,Decision curve analysis (net benefit),Determinants of prediction correction (conference minus formula)	—

Countries

Japan

Contacts

Public ContactAkihiro Jino

Morinomiya Hospital Neurorehabilitation Research Institute

jino@omichikai.or.jp06-6969-0111

Outcome results